The long-term objective of the proposed research is to understand how the internal circuits of mammalian retinas shape the receptive fields and response properties of the retinal ganglion cells. The specific aim is to complete the census of the rabbit retina's major cell populations. Mammalian retinas have been known for a long time to contain many cell types and subtypes. Although it is unlikely that we can analyze all of the retina's rare (infrequent) cells, we believe that we are now able to identify and map all of the major ones. The initial question is: what are the total numbers of horizonal, bipolar, and amacrine cells? The question will be approached by two independent sets of experiments. In the first, populations of retinal cells will be sequentially identified by staining. Generally available stains, fluorescent labels and antibodies will be used to distinguish individual classes of cell. In the second, the three major classes of retinal cells will be distinguished by confocal microscopy. The distribution of each cell type across the retinal surface will be mapped quantitatively. These experiments will make available (for the first time) an estimate of the total number of amacrine cells. By comparing that number with the total numbers of the amacrine cells that have been stained by histochemical and autoradiographic methods, we will learn how many unidentified amacrine cells must still exist. The shapes of the unidentified cells will then be sought by experiments in which amacrine cells are injected with lucifer yellow. Injection will be combined with fluorescent, histochemical or autoradiographic labeling of cell bodies to give the population densities of the injected cells.